Evidence (8807 claims)
Search and filter individual claims pulled from the papers. Looking for a specific finding ("what's the effect on wages?"), you're in the right place. Want to compare whole outcome categories against each other instead? Use the Evidence Explorer.
The board below groups claims two ways: by broad theme (nine paper-level topics) and by outcome category (the 34 claim-level outcomes that the Explorer and Syntheses also use).
Browse by theme
Nine broad, paper-level topics. Click one to filter the claims below.
Adoption
9875 claims
Filter claims →
Productivity
8807 claims
Filtered →
Governance
7870 claims
Filter claims →
Human-AI Collaboration
7560 claims
Filter claims →
Org Design
4892 claims
Filter claims →
Innovation
4781 claims
Filter claims →
Labor Markets
4004 claims
Filter claims →
Skills & Training
3308 claims
Filter claims →
Inequality
2332 claims
Filter claims →
Claims by outcome category
Counts by direction of finding. These are the same 34 outcome categories the Explorer compares and the Syntheses are written for. A linked row has a published synthesis.
| Outcome | Positive | Negative | Mixed | Null | Total |
|---|---|---|---|---|---|
| Other | 870 | 233 | 116 | 1066 | 2363 |
| Governance & Regulation | 976 | 451 | 218 | 133 | 1809 |
| Organizational Efficiency | 949 | 224 | 144 | 88 | 1416 |
| Technology Adoption Rate | 764 | 287 | 141 | 122 | 1325 |
| Research Productivity | 501 | 152 | 74 | 362 | 1101 |
| Output Quality | 542 | 216 | 69 | 69 | 896 |
| Decision Quality | 387 | 198 | 94 | 54 | 740 |
| Firm Productivity | 513 | 67 | 101 | 27 | 714 |
| AI Safety & Ethics | 249 | 303 | 73 | 36 | 667 |
| Market Structure | 190 | 192 | 134 | 27 | 548 |
| Task Allocation | 243 | 77 | 91 | 36 | 452 |
| Innovation Output | 291 | 33 | 55 | 20 | 401 |
| Skill Acquisition | 206 | 72 | 65 | 21 | 364 |
| Employment Level | 133 | 63 | 115 | 22 | 335 |
| Fiscal & Macroeconomic | 153 | 79 | 52 | 32 | 323 |
| Task Completion Time | 206 | 37 | 12 | 15 | 272 |
| Firm Revenue | 179 | 52 | 29 | 5 | 266 |
| Consumer Welfare | 130 | 76 | 47 | 13 | 266 |
| Inequality Measures | 48 | 137 | 51 | 6 | 242 |
| Worker Satisfaction | 101 | 81 | 25 | 13 | 220 |
| Error Rate | 84 | 110 | 11 | 5 | 210 |
| Wages & Compensation | 98 | 47 | 30 | 10 | 185 |
| Regulatory Compliance | 88 | 73 | 17 | 7 | 185 |
| Automation Exposure | 66 | 64 | 33 | 16 | 182 |
| Team Performance | 105 | 29 | 30 | 11 | 176 |
| Training Effectiveness | 109 | 22 | 14 | 21 | 168 |
| Developer Productivity | 114 | 21 | 14 | 8 | 158 |
| Job Displacement | 12 | 90 | 24 | 1 | 127 |
| Hiring & Recruitment | 57 | 9 | 9 | 5 | 80 |
| Skill Obsolescence | 6 | 56 | 9 | 1 | 72 |
| Social Protection | 43 | 17 | 8 | 2 | 70 |
| Creative Output | 35 | 21 | 9 | 4 | 70 |
| Labor Share of Income | 18 | 21 | 17 | 1 | 57 |
| Worker Turnover | 15 | 16 | — | 4 | 35 |
| Industry | — | — | — | 1 | 1 |
Productivity
Remove filter
Current agents are not yet able to reliably produce professional-quality spreadsheets at the level of complexity real-world workflows demand.
Conclusion/interpretation based on the benchmark results and qualitative review reported in the paper (supporting quantitative details not shown in the excerpt).
Even the strongest agents frequently fall short of professional finance standards and degrade sharply as the difficulty increases beyond a few chained calculations.
Empirical observation from the paper's benchmark showing performance declines with task difficulty (no numeric breakdown or sample sizes provided in excerpt).
Existing spreadsheet benchmarks do not measure this advanced capability, focusing instead on question-answering or single-formula edits.
Paper's review/assessment of prior benchmarks (stated in introduction); no comprehensive benchmark list or counts provided in excerpt.
Seventy-four percent of task misalignments could be attributed to developers who tended to overfocus on efficiency and speed, especially for systems performing tasks in people-facing occupations such as those in the human resources sector.
Result from comparing traits causing incidents to developers' stated preferences (sample of 197 developers) and computing the proportion of misalignments where developer-desired traits matched the traits causing incidents; noted sectoral concentration in people-facing occupations (e.g., HR).
In most cases, workers wanted systems that are precise, insightful, or personal, but instead received systems that are basic, simple, or general.
Qualitative/quantitative comparison of preferred traits (from 202 workers) versus traits observed in AI systems in incident reports (LLM-coded); reported dominant preference traits versus dominant delivered traits.
As many as 83% of workplace incidents stem from worker-AI misalignments.
Result from comparing LLM-extracted traits of AI systems (from 1,524 incident reports) to the traits preferred by workers (sample of 202); counted incidents where traits did not match worker preferences and reported proportion.
Commercial demand drivers systematically distort finished-goods inventory targets and require integration with sales-and-operations planning for accurate calibration.
Narrative synthesis of studies addressing demand-driver effects on finished-goods targets and recommendations for S&OP integration.
4.7% of modified files introduce new Bandit findings (security issues).
Static security analysis using Bandit run before and after each refactoring change on the AIDev Python PRs; reported proportion of modified files that gained new Bandit findings.
24.17% of modified files introduce new Pylint issues, predominantly convention-level violations such as long lines.
Domain-independent static analysis using Pylint applied before and after refactoring commits in the AIDev Python PRs; proportion of modified files with new Pylint findings reported.
However, models remain limited in long-horizon reliability and domain-specific planning.
Evaluation results and analysis in paper highlighting failures in maintaining reliability over long-horizon tasks and in planning for domain-specific workflows.
Extensive evaluations reveal that existing agents achieve only 36.0% task success on realistic media editing tasks.
Empirical evaluation reported in paper measuring task success rates of existing GUI agents on the Cutverse benchmark (benchmark size: 186 tasks across 7 apps implied).
There is a "Sim-to-Real" gap: synthetic tests maintain constant memory usage but realistic workflows exhibit linear memory growth of about 3 tokens per message, with consolidation quality emerging as the primary scalability bottleneck.
Empirical comparison between synthetic tests and realistic workflows over the reported message corpus (15,000 messages); reported growth rate (~3 tokens/message) and qualitative identification of consolidation quality as bottleneck.
Full-context models fail at 10,000 messages due to context overflow.
Empirical comparison reported in the paper (within the large-scale evaluation), statement that full-context models fail at ~10,000 messages.
Monolithic approaches suffer from quadratic cost scaling and cognitive degradation when used for long scientific workflows.
Author statement contrasting monolithic approaches with the proposed architecture; conceptual/architectural claim rather than a single quantified experiment.
Context window saturation is a critical bottleneck as LLMs evolve into persistent scientific collaborators, because iterative data analysis and hypothesis refinement rapidly saturate even extended contexts with dense technical content.
Author claim based on observed behavior of scientific workflows; contextual motivation in paper (no specific experiment cited for this general statement).
In those same benchmarks, 16 of 84 tasks suffered negative deltas when Skills are introduced.
Reference to the same prior benchmark aggregation that reported task-level deltas (count of tasks with negative deltas = 16 out of 84).
Underspecified prompts can lead to low-quality answers and additional interaction.
Motivating claim in the paper; presented as the problem the study addresses (no sample size or statistical test reported in the abstract).
Recent generative models show promise, yet they lack explicit mechanisms to balance exploration and safety, relying solely on action perturbations or trajectory guidance without a safety fallback, resulting in inefficient exploration and elevated financial risk for advertising platforms.
Argument in the paper contrasting generative-model-based approaches with the authors' proposed solution (conceptual claim; no quantitative backing given in the excerpt).
Reinforcement Learning approaches modeled bidding as a Markov Decision Process but struggled with long-term dependencies.
Statement in the paper summarizing limitations of prior RL-based bidding work (qualitative claim; no experimental details or sample size provided in the excerpt).
Early rule-based methods lacked adaptability.
Literature/contextual statement in the paper's introduction summarizing prior approaches to automated bidding (no empirical data or sample size reported).
Twin agents dissolve that boundary, raising a class of trust calibration challenge these frameworks were not designed to handle.
Argument and design observations from the authors' ongoing project presented in the paper; conceptual claim explaining why existing frameworks may be insufficient for twin agents.
When a human colleague doubts a twin agent's output, they face three failure modes (a schema gap, an epistemic gap, and a model artifact) with no reliable attribution path between them.
Conceptual taxonomy derived from the authors' early design observations; presented as an identified set of failure modes in the paper (qualitative, no numeric sample reported in abstract).
Drawing on early design work in an ongoing project, we identify a trust calibration problem specific to this approach.
Based on the authors' early design work (qualitative/design research) described in the paper; no sample size or quantitative metrics reported in the abstract.
Frontier agents struggle with end-to-end completion despite partial progress.
Evaluation experiments reported in the paper showing frontier (state-of-the-art) agents achieving partial progress but failing to reliably complete end-to-end tasks in the OpenComputer benchmark.
Major open challenges for responsible adoption include reliability, bias, privacy, automation bias, transparency, and evaluation.
Authors' identification of risks and open research challenges based on their review/analysis (conceptual synthesis).
Current AI support for code review remains fragmented, with tools focusing on isolated tasks such as reviewer recommendation, PR description generation, or comment suggestion rather than the end-to-end PR review workflow.
Authors' survey/overview of existing AI tooling for code review described in the paper (conceptual / review-based evidence). No quantitative counts provided in the abstract.
AI coding assistants expand the volume of code requiring review, turning code review into a growing bottleneck.
Authors' analytical claim linking increased code production from AI assistants to increased review workload; presented as an observed/trend claim in the paper rather than supported by a quantified study in the abstract.
Code review has evolved for decades, from informal peer checking to today's pull request (PR) workflows, yet it remains a largely manual, uneven, and cognitively demanding process.
Authors' literature review and historical synthesis of code review practices presented in the paper (conceptual / review-based evidence). No empirical sample or experiment reported in the abstract.
Challenges including algorithmic bias, data privacy concerns, high costs, and skill gaps persist across contexts.
Cross-study synthesis of barriers and challenges reported in the 21 included studies spanning multiple contexts.
SMEs face unique resource constraints yet lag in AI-HRM adoption.
Synthesis conclusion from the systematic review of 21 included studies (published 2019–2026) comparing adoption patterns and barriers for SMEs.
Greater automation can obscure rather than eliminate failure modes.
Analytical claim in paper arguing that increased automation hides failures; presented as an interpretive finding rather than a quantified experimental result in the excerpt.
End-to-end autonomous systems have not yet consistently reached major-venue acceptance standards.
Paper's statement based on review of acceptance/peer-review outcomes and standards as of April 2026; no numeric acceptance-rate data presented in the excerpt.
Research code lags far behind pattern-matching benchmarks.
Paper's evaluative claim from its experiments/coding analysis indicating code produced for research tasks is weaker than benchmark performance on pattern-matching tasks; excerpt contains no numerical comparison.
Generated ideas often degrade after implementation.
Paper statement about the gap between idea generation and implemented results reported in the Creation-phase analysis; no quantified follow-up study reported in the excerpt.
AI remains fragile for genuinely novel ideas, research-level experiments, and scientific judgment.
Summary claim from the paper's end-to-end lifecycle analysis indicating limitations on novelty and experimental rigor; no numeric performance metrics provided in excerpt.
Frontier LLMs fail to judge novelty reliably.
Paper's claim from its Validation-phase analysis that models do not reliably assess novelty; excerpt contains no underlying experimental sample or validation metrics.
Frontier LLMs miss hidden errors.
Qualitative statement from paper indicating models fail to detect some latent or subtle errors in research artifacts; no numeric evaluation provided in excerpt.
Under scientific pressure, even frontier LLMs still fabricate results.
Reported observation in paper about model behavior under scientific-use conditions; no specific quantitative experiments or sample sizes given in the excerpt.
Diagnostics also reveal a small tail of extreme errors for the Random Forest model.
Model diagnostic analyses reported in the paper indicating error distribution and presence of extreme prediction errors (tail).
Unrestricted frontier-scale checkpoint synthesis remains open (i.e., not yet solved).
Authors' assessment in the abstract noting current limits; asserts that unrestricted synthesis at frontier/model-scale has not been achieved.
Real-world trajectory data can provide highly accurate insights but collecting it is costly and often infeasible for many retailers.
Author claim about practical constraints of data collection for retailers; argued contextually in the paper rather than presented as a quantified empirical finding in the excerpt.
Actual customer trajectories deviate by an average of 28% from shortest paths.
Empirical measurement reported in the paper comparing real-world trajectory data to shortest-path (TSP-like) routes; exact sample size not stated in the provided text.
In the context of search retrieval, current cold-start models suffer from the misalignment between training objectives and online business metrics, and they lack effective mechanisms to measure an item's growth potential.
Claim made in paper as motivation/background; no empirical details provided in the excerpt.
Existing systems tend to prioritize presenting users with already popular items, a phenomenon often referred to as the "Matthew effect".
Statement/observation in the paper; presented as background/motivation (no empirical evidence or sample size reported in the excerpt).
AI agents deployed into SRE workflows currently derive their understanding of environment state from raw observability telemetry at query time, paying a semantic-interpretation tax in tokens, latency, and inferential reliability.
Author statement / problem framing in the paper (no quantitative experiment reported for this general claim).
Standard health system digital transformation policy, which typically addresses only the threshold failure through individual incentives, is predicted to systematically produce the partial adoption trap.
Model prediction contrasting full policy architecture vs. conventional policies that focus solely on individual incentives; analytical conclusion that such limited policies leave other failure modes unaddressed and therefore lead to stable partial adoption. Theoretical model; no empirical sample.
The barrier-lowering benefit of failed attempts is offset when trust erosion is rapid.
Model analysis combining cost-ratchet dynamics and trust erosion parameters; results showing interaction where fast trust erosion negates barrier reductions. Theoretical simulations/derivations; no empirical sample.
These failure modes are most severe precisely for the technologies with the greatest systemic value: the Value-Adoption Paradox.
Analytical result from the model showing failure-mode severity as a function of systemic value; theoretical identification of a paradox where higher systemic-value technologies face stronger coordination/trust/cultural barriers. Theoretical derivation; no empirical sample.
The basin of attraction of the partial adoption trap is enlarged by a cultural failure arising from negative coordination norms among doctors.
Model analysis including cultural coordination norms; theoretical demonstration that negative norms exacerbate partial adoption equilibria. Theoretical model; no empirical sample.
The basin of attraction of the partial adoption trap is enlarged by a trust failure arising from the organisation's inability to credibly commit to sharing productivity gains.
Model extension incorporating organisational commitment/transfer of gains; analytical results showing trust/commitment constraints increase stability of partial adoption. Theoretical model; no empirical sample.